Web Analytics

The main issue concerning web analytics is that we get large quantities of data, but no insights (information – data with meaning attached to it). The raw data is referred to as clickstream data. This click-level data needs to be interpreted to really bring some value. Clickstream data tells us the ‘what’, but we need to get the insight – ‘why’. In order to get to the ‘why’, we have to work our way through several stages.

The first stage after gathering the clickstream data is a multiple outcomes analysis. That means we have to tie the web-site outcomes (increased revenue, reduced costs, improved CRM) to all the activities we do on our site so that these activities support our multiple outcomes. The next stage is experimenting and testing. That means we can use tools such as Google Website Optimizer or learn from our customers through feedback. Then we can adjust our site and see if the click-level data improves. This stage already partially tells us the ‘why’, because we leverage opinions of our visitors. The next stage fully utilizes the voice of customer. We get a clear picture of why we get the clickstream data. It is possible to implement another stage, which deals with competitive intelligence. Through certain web tools, we can compare our clickstream data to data of our competitors.



Web analytics are usually presented either in a form of metrics or as a KPI – key performance indicator. A metric is a quantitative measurement describing data and trends, whereas KPI is a report explaining how we are doing against our objectives. When someone comes to our site, we call him a visitor. If the same subject comes again and doesn’t block the first party cookies, we can identify him as a unique visitor. The whole time spend at all the pages of our site is then called a visit.

There are six foundation metrics – visits, bounce rate, page views, pages/visit, average time on site and % of new visits. The one that probably needs to be explained further is the bounce rate. It is the amount of visitors, who left the site instantly without any further clicking.

There are several visitor acquisition methods – direct traffic (visitor searches directly for your URL), search engine, other (banner ads…) and referring sites.

Hubbard One came up with a new web analytics tool, which is designed specifically for law firms. Apparently, law firms don’t have the skill and time to understand analytical data that isn’t tailored to their needs and so Hubbard One decided to fill this gap.

The privacy issue of web analytics is a very hot and discussed topic. A web analytics firm and its clients, including about.me and Spotify, were sued for crossing the legal line of gathering data about their browsing customers. They attached a new set of cookies to these customers’ browsers even though the customers didn’t want to be tracked, and therefore unattached the previous cookies.